New Orleans drinking water sources tested by gas ... - ACS Publications

Aug 1, 1975 - New Orleans drinking water sources tested by gas chromatography-mass spectrometry. Occurrence and origin of aromatics and halogenated ...
0 downloads 0 Views 426KB Size
New Orleans Drinking Water Sources Tested by Gas Chromatography-Mass Spectrometry Occurrence and Origin of Aromatics and Halogenated Aliphatic Hydrocarbons Betty J. Dowty, Douglas R. Carlisle, and John L. Laseter* Department of Biological Sciences, University of New Orleans, Lakefront, New Orleans, La. 70122

Low-molecular-weight volatile organics were thermally extracted from water samples before, during, and after processing by a New Orleans area municipal water treatment facility. The volatile components were trapped on a solid polyphenyl ether adsorbent and resolved by high resolution gas chromatographic procedures. A gas chromatographmass spectrometer-computer combination was used to identify major aromatic and halogenated aliphatic components. Approximately 60-70 organics were commonly present in finished water. They varied considerably from day to day in their relative concentrations. A commercial source of artesian water, as well as deionized charcoal-filtered finished water, was also analyzed. These latter water sources contained a wide variety of organics of which many are common to finished water.

There are several techniques, such as liquid-liquid solvent extraction techniques ( I ) , recovery on ion-exchange resins such as Amberlite ( I ) , carbon adsorption (2), and head space sampling techniques (3, 4 ) to extract organics from water for analysis. A novel technique described herein has been successfully applied to the qualitative analysis of trace volatile organics in water following their thermal extraction. Previous reports by this laboratory have employed this new technique to monitor low-molecular-weight organics present in blood plasma of New Orleans residents and their drinking water ( 5 ) , and body tissues (6). These studies revealed many halogenated hydrocarbons in New Orleans drinking water, several of which had not previously been identified. Another study conducted to locate the origin of these halogenated hydrocarbons (7) revealed that some of the compounds appear to be introduced in the water treatment process itself, a t the chlorination step of the procedure. This is not the first time that the chlorination process has been suspect. Independent studies by another laboratory also indicate that it is likely that chlorination introduces new compounds into the water (8), although identification of these compounds was incomplete. Because little is known of the origin of these compounds, this study was undertaken to identify by mass spectrometry the low-molecular-weight organics isolated at various stages of the water treatment process. Finished water used in this investigation was derived from the lower Mississippi River. Two alternate sources-i.e., deionized charcoal-filtered water, and bottled artesian water were also studied. Major emphasis was placed on the identification of halogenated aliphatic and aromatic constituents of water, as these are the compounds of interest from a health standpoint. Little is known of the toxic and health aspects of these compounds when ingested by man over long periods of time. Even less is known about antagonistic or synergistic effects elicited by these compounds acting in the presence of one another. 762

Environmental Science & Technology

Experimental Sample Collection and Source Description. Water samples were collected and stored in l-liter acid-washed glass bottles with screw caps until analyses were performed. One-liter water samples were collected for chromatographic analyses, whereas approximately 3-liter samples were collected for mass spectrometric analyses. Samples representative of various steps in the water treatment process were collected from the local treatment plant that receives its water supply from the lower Mississippi River. Water treatment procedures a t this plant were those common to solid contact, up-flow type units. The water is initially treated with several chemicals to control alkalinity, soften the water, and oxidize organics present. Next, the water sample is treated to remove colloidal and larger size particles at the clarifier stage of the water treatment process. En route from the clarifier to the sand and gravel filter beds, chlorine and ammonia are added to the water. In this latter process mono- and dichloramines are formed. Water samples were analyzed by gas chromatographicmass spectrometric methods from the treatment plant entrance, the clarifier effluent (which is after sedimentation and some chemical treatment, but before chlorination), and the finished water (tap water). Bottled artesian water from a local supplier was also analyzed, as well as water from a commercial laboratory charcoal filtering-deionizing unit. The commercial filtering unit (Continental Deionized Water Service, New Orleans, La.) is comprised of 0.0070 meter3 of 18-40 mesh, grade 718 charcoal contained in a glass fiber tank, as well as Amberlite IR 120 cation-exchange and Amberlite IRA 410 anion-exchange resins in a 2-3 ratio. Sample analyses were performed according to the technique described earlier (5, 7). Volatile organics were eluted from the water sample by heating to 95°C under a stream of ultrapure helium. Totally stainless steel valves and regulators (Matheson, LaPorte, Tex.) ultrasonically cleaned, were incorporated into the helium flow system. Additionally, one molecular sieve 5A (Linde, Applied Science, State College, Pa.) and two activated charcoal (Adsorbit, Curtin, New Orleans, La.) filters were employed. The helium stream was passed over the water sample contained in a 1liter round bottom flask and exited through a series of glass condensers to eliminate the bulk of water vapor. Volatile organics were trapped for 1 hr on poly p-2,6-diphenyl phenylene oxide adsorbent having 35-60 mesh size (Applied Science, Laboratories, Inc., State College, Pa.) attached to the end of a condenser train. After the polymer was trapped, it was transferred to a glass injection port liner. The injection port of a Hewlett-Packard 7620-A gas chromatograph has been modified to accommodate the larger diameter liners. The liners containing the trapped organics were placed into the injection port maintained at 200OC. A helium stream of approximately 16 ml/min was used to transfer the volatile components onto a 1.5 m X

Table I. Organic Constituents Identified by Gas Chromatography-Mass Spectrometry in Mississippi River Clarifier Effluent and Finished Water Relative concentration$

Chromatographic peak no.

Compounds identified

Finished

Clarifier effluent

1 5 6 9 14 21 22 26

Acetaldehyde Acetone Diethylether tert-Bu ty I a I co h o I Dichlorornethane Dichioroethylene + tetrahydrofuran Chloroform Benzene, carbon tetrachloride, dichloroethane 3-Methyl butan-1-al Trichloroethylene + dichloropropane Dichloropropene Brornodichlorornethane Toluene

1.28 7.29 0.77 4.18 0.33 0.64 8.73 36.61

1.38 0.96 0.34 3.11 0.25 1.27 1.87 3.26

1.11 8.28 0.29 12.31 1.51 0.51 1.02 0.3 1 2.48 0.98 0.29

0.20 11.48 0.34 38.46 0.57 0.20 2.03 0.64 18.16 0.90 0.47

28 31 32 34 37 39 43 44 49 58 65

CEH,,

Trichloroethane Tetrachloroethylene Di bromochloromethane (2- or 1-Napthy1)dichlorornethane Xylene

Mississippi River

2.10 5.60 2.20 11.13 0.77 0.93 1.17 12.27

20.87 1.00 2.00 0.67 5.67 1.87 2.17 0.80 0.60

aAs c o m p u t e d from gas c h r o m a t o g r a p h i c d a t a

MISS RIVER

-

5. 43k

I

-

23

34

49

-58

3!

--\-

3, 3?

-

65

-

- _-

,

CLARIFIER EFFLUENT

49,

9

? 10 I

5

36

23 15 21 22

4:

59

37

65

i

"_

~

5

2 2 2:

9

I

c-

-

+

FINWED WATER

31 34

?

49

I

I

37;

23

36

9

2, $20

43 59

28 33

,?

44 ~

-

$5

x _ , i

- " ._-

BLPNK

-25-, -~ ~~~

10

45 20

125

85 30

40

--

50

€0

~

K5

70

80

T I M (MINUTES1

chromatographic separations of volatile organics collected from one-liter samples of Mississippi River water at the purification plant entrance (upper trace), the clarifier effluent stage (second trace), and the finished water (third trace). A glassware and gas chromatographic system blank is also shown (lower trace). Chromatographic conditions are described in the text. Identification of numbered peaks is given in Table I Figure 1. Gas

0.05 cm precolumn coated with either Emulphor ON-870 or GE SF-96 and Igepal CO 880 (Applied Science Laboratories, Inc.) held in a Dry Ice-methanol bath. After a 7-min trapping period, the cold bath was removed and the collected organics were swept onto a capillary chromatographic column (9). All elassware was cleaned with chromic acid solution and purged with an ultrapure gas stream for 1 hr while being held a t a temperature of 165OC. This operation will tend to remove residual traces of organics from previ~ U analyses S and thereby reduce background. Blanks were run immediately before a given analysis to ensure that the helium carrier gas, glassware, and chromatographic system were free of any interfering organic volatiles. Mass spectrometric analyses were performed on 3-liter water samples by venting the gas chromatographic effluent from the end of the column through a heated (200°C) stainless steel transfer line into the ion source of a Du Pont 21-491 double-focusing mass spectrometer. The source was maintained at 21OOC and an ionizing voltage of 70 eV was used. The signal from the electron multiplier amplifier was processed by a PDP-12 LDP (Digital Equipment Corp., Maynard, Mass.) computer for analysis and data reduction.

Results and DiscussLon Figure 1 illustrates the typical gas chromatographic separation of organic constituents present in the Mississippi River water a t the plant entrance (upper trace), the clarifier effluent stage of the water treatment process (second trace), the finished water (third trace), and a glassware and chromatographic system blank (lower trace). Mass spectrometric identification of these compounds and their relative percent abundances in the chromatograms are presented in Table I. Of the approximately 70 organic constituents present in finished water that can be detected in any single analysis of liter size, the preponderance of organics present is either aromatic or halogenated aliphatic and aromatic. As can be seen, the major constituents of finished water are benzene, carbon tetrachloride, and dichloroethane (which remained unresolved in peak 26), bromodichloromethane Volume 9, Number 8,August 1975 763

ARTESIAN WATER W

v,

;2:

4!'

1

\r

1-25-

10

4g 20

85 30 40 TIME (MINUTES)

195

125 60

50

70

80

Figure 2. Gas chromatographic separation of volatile organics collected from a commercial source of bottled artesian water. Chromatographic conditions are described in the text. Identification of numbered peaks is given in Table II (peak 34), and chloroform (peak 22). Many halogenated and aromatic organics such as trichloroethylene (peak 44), (2- or 1-naphthyl)-dichloromethane(peak 58), and xylene (peak 65) pass through the treatment plant unaltered in concentration. By contrast, however, dichloroethylene (peak 21), chloroform (peak 22), benzene, carbon tetrachloride and dichloroethane (peak 26), bromodichloromethane (peak 34), toluene (peak 37), and dibromochloromethane (peak 49) are enhanced greatly in the water treatment process. Rise in concentrations of these compounds after chlorination strongly suggests they are formed by chlorination of precursor organic molecules. Many aromatic compounds, as well as bromine-containing compounds, also appear elevated after the chlorination procedure. This strongly implies that any accidental or industrial pollution does have a direct bearing on the quality of the finished water. I t should be noted that essentially every compound observed in finished water was present, usually in a much lower quantity, in the precursor river water. Previous work (7) indicates that many compounds generated in the purification process are discharged into the river. One may assume that the presence of trace quantities of such compounds may have originated from water treatment plant discards occurring upstream. Many of the halogenated and aromatic

Table II. Organic Constituents Identified in Commercially Bottled Artesian Water Chromatographic peak no.

6 7 12 20 23 28 33 39 41 51 53 59 61 66 79

Compds identified

Relative0 concns

Acetone Diethy lether Dichloromethane Ethylacetate Chloroform Benzene Dichloropropane Bromodichloromethane Toluene Dibromochloromethane Xylene

1.22 2.79 0.96 1.31 5.72 16.16 0.26 0.87 6.11 0.48 1.97 0.66 0.13 1.22 2.40

C,Hm C,oHzz C,,H*4

Dichlorobenzene

a As c o m p u t e d f r o m gas c h r o m a t o g r a p h i c data.

764

Environmental Science & Technology

compounds present in the river water are still present in the finished water (Table I). This suggests that they pass through the treatment plant without removal. A commercial source of bottled artesian water was also examined, as it serves as an alternative to drinking finished I

3

59

C W K M L FILTERED

!

DEIONIZED WATER

74

I c25-+ 10

45

-

20

30

05 40

'

50

125 60

70

65 80

-__ 90

T I M IMWTESI

Figure 3. Gas chromatographic separation of volatile organics collected from finished water following treatment by a commercial deionizing-charcoal filtering unit. Chromatographic conditions are described In the text. Identification of numbered peaks is given in Table 111

Table 111. Organic Constituents Identified in Commercial Deionized Charcoal-Filtered Water

Chromatographic peak no.

1 3 4 9 19 23 28 30 32 42 50 56 57 58 63 64 74 89

Compds identified

Acetaldehyde Acetone Diethylether tert-Butylalcohol C h I orof o rm Benzene Trichloroethylene C,H,,

Methylmethacry late Toluene Tetrachloroethylene Monochlorobenzene Ethyl benzene Xylene Xylene Styrene Methylethyl benzene Dichlorobenzene

a As c o m p u t e d f r o m gas c h r o m a t o g r a p h i c data.

Relativea concns

2.71 7.43 0.91 1.09 1.00 2.17 1.74 1.00 0.46 3.00 13.77 0.89 1.94 2.31 0.71 3.86 6.00 1.94

water in the local area. The chromatographic results are shown in Figure 2. Compound identification and the relative percent abundance are shown in Table 11. By examining Figure 2, it is evident that there are much lower concentrations of organics in the artesian water. However, it can be seen that many of the organics are the same as those in the finished water. In this particular sample, chloroform, benzene, and toluene are the major constituents. Toluene concentrations appear to be enhanced over the concentration in finished water. Variations in organics present and their relative concentration varied from container to container of artesian water from the same commercial supplier. Because charcoal filtering is currently being suggested as a means to eliminate the problem of organics in water, a study was undertaken to evaluate the efficiency in removal of organics in this process. Water from a commercial deionizing-charcoal filtering unit was analyzed, and its chromatogram is depicted in Figure 3. Table I11 identifies the compounds, their gas chromatographic peak number, and their relative percent abundance. Chloroform, benzene, and toluene are still present in the water. Chloroform and benzene show reduction in concentration over those amounts present in finished water; toluene, on the other hand, appears to have increased somewhat. Tetrachloroethylene shows a marked increase over the amount found in finished water or commercial sources of bottled water. Of interest is the appearance for the first time of compounds such as methylmethacrylate and styrene. These

compounds could have originated from the plastics used somewhere in the preparation or storage of the ion-exchange resins or charcoal. Our preliminary data seem to indicate that further studies are needed to determine the value of charcoal filtering before its incorporation on a large scale into water treatment plants.

Acknowledgments The authors would like to acknowledge the cooperation of N. Brodtmann, Jr., in obtaining the water samples. We also acknowledge the technical assistance of R. Evans, M. Legendre, and D. Condie. Literature Cited (1) Burnham, A. K., Calder, G. V., Fritz, J. S., Junk, G. A,, Svec, H. J., Vick, R., J . Am. W a t e r Works Assoc., 65, 722-25 (1973). (2) Grob, K., J . Chromatogr., 84,255-73 (1973). (3) Novak, J., Zluticky, J., Kubelka, V., Mostecky, J., ibid., 76, 45-50 (1973). (4) Mieure, J . P., Dietrich, M. W., J . Chromatogr. Sci., 11, 559-70 (1973). (5) Dowty, B., Carlisle, D., Laseter, J. L., Science, 187, 75-7 (1975). (6) Politzer, I. R., Githens, S., Dowty, B., Laseter, J. L., J. Chromatogr. Sci., accepted 1975, in press. (7) Dowty, B., Laseter, J . L., Anal. Lett., 8 (l), 25-32 (1975). (8) Glaze, W. H., Henderson, J . E., IV, Bell, J. E., Wheeler, V. A., J . Chromatogr. Sei., 11,580-4 (197B). (9) Mon, T. R., Res. Deu., 14 (1971).

Received for review December 12,1974. Accepted April 21, 1975.

NOTES

Trace Elements in Corn Grown on Long-Term Sludge Disposal Site M. B. Kirkham’ Plant Physiologist, Advanced Waste Treatment Research Laboratory, Environmental Protection Agency, Cincinnati, Ohio 45268

Corn plants (Zea mays L.) grown on soil treated with municipal sludge (28 metric tons/ha/yr) for 35 years were analyzed for Al, Cd, Cu, Fe, Mn, Ni, Pb, and Zn. Soil was analyzed to the 91-cm (36-in.) depth for total and dilute acid extractable concentrations of the trace elements. Total concentrations of the trace elements in the 0- to 30-cm (0- to 12-in.) depth of soil were increased by sludge additions by the following factors: Al, 1.1; Cd, 35; Cu, 16.5; Fe 1.1; Mn,

1.2; Ni, 2; Pb, 16.5; Zn, 13. The results indicated that, even though large amounts of trace elements accumulated in the soil, these elements did not accumulate appreciably in the grain of the corn plants when compared with the accumulation in the control crop and normal levels reported in the literature. The roots, however, contained large amounts of the elements. Concentrations of Cd and Cu in the leaves were higher than normal concentration ranges.

Sludge has been applied to land for decades in the USA for disposal purposes and to use it as a soil conditioner and fertilizer ( 1 ) . Long-time spreading of sludge on land, however, can result in high concentrations of trace elements in soils because many metals present in sewage are removed by sewage treatment and are concentrated in the sludge (2). There is concern that plants grown on these soils may accumulate trace elements in amounts toxic to man and animals eating the crop. Almost no information has been published on the concentration in soils of available trace elements from sludge being applied to land over many years. Most studies on metal concentrations in sludge-treated

soils have been started within the last five years ( 3 , 4 ) .This paper reports the concentrations of trace elements in corn plants and soil treated with sludge for 35 years.

Present address, Department of Plant and Soil Sciences, University of Massachusetts, Amherst, Mass. 01002.

Materials and Methods Corn plants (Zea mays L.) were sampled on October 19, 1973, a t the Dayton, Ohio, Sewage Treatment Plant. The soil type a t the plant is Warsaw silt loam ( 5 ) .The amount of erosion is slight (slope of land, 0-2%). The well-drained soil with moderate permeability is over limestone bedrock and has formed in alluvium of calcareous glacial till. Two plant samples were obtained from each of three different locations: a control plot (1.2 ha) that had never received sludge; a plot (0.2 ha) that had been saturated with supernatant from sludge-holding lagoons between 1968 and Volume 9, Number 8, August 1975

765